ATM signalling facilitates repair of DNA double strand breaks
associated with heterochromatin
Aaron A. Goodarzi1, Angela T. Noon1, Dorothee Deckbar2, Yael Ziv3, Yosef
Shiloh3, Markus Löbrich2, Penny A. Jeggo1*
1Genome Damage and Stability Centre, University of Sussex, East Sussex BN1
9RQ, United Kingdom
2Darmstadt University of Technology, Radiation Biology and DNA Repair, 64287
3The David and Inez Myers Laboratory for Genetic Research, Department of
Human Molecular Genetics and Biochemistry, Sackler School of Medicine, Tel
Aviv University, Tel Aviv 69978, Israel.
*Corresponding author P.A.Jeggo@sussex.ac.uk
Characters (excluding spaces): 44,966
Running Title: ATM facilitates DSB repair within heterochromatin
Keywords: ATM / KAP-1 / Heterochromatin / DSB repair / Ionizing Radiation
Subject Category: Genome Stability and Dynamics
ATM-signalling is essential for the repair of a subset of DNA double strand
breaks (DSBs); however its precise role in DSB-repair is unclear. Here, we show
that no more than ~25% of DSBs require ATM-signalling for repair and this
correlates with increased chromatin complexity but not damage complexity.
Importantly, we demonstrate that heterochromatic DSBs are generally repaired
more slowly than euchromatic DSBs and that ATM-signalling is specifically
required for DSB-repair within heterochromatin. Significantly, knockdown of the
transcriptional repressor KAP-1, an ATM substrate, or the heterochromatin-
building factors HP1 or HDAC1/2 alleviates the requirement for ATM in DSB-
repair. We propose that ATM signalling temporarily perturbs heterochromatin via
KAP-1 and that this is critical for DSB-repair/processing within otherwise
compacted/inflexible chromatin. In support of this, ATM-signalling alters KAP-1
affinity for chromatin enriched for heterochromatic factors. These data suggest
that the importance of ATM-signalling for DSB-repair increases as the
heterochromatic component of a genome expands.
DNA double strand breaks (DSBs) lead to chromosomal fragmentation
and genomic rearrangements if not repaired in an accurate and timely manner. In
mammaliancells,DSBstriggera signallingresponse from theAtaxia
Telangiectasia (A-T) Mutated (ATM) pathway while DNA non-homologous end
joining (NHEJ) operates as the primary mechanism of repair. Whilst the majority
of DSB repair is ATM-independent, loss of ATM or DNA damage response
mediator proteins required for ATM signalling (such as the Mre11/Rad50/NBS1
(MRN) complex, 53BP1 or histone H2AX) results in a defect characterized by
~10-25% of initially incurred DSBs remaining un-repaired in non-dividing cells
(Riballo et al., 2004; Deckbar et al., 2007). This defect, although subtle, confers
profound radiosensitivity to A-T cells (Riballo et al., 2004); notably, ATM mutation
is associated with one of the highest degrees of radiosensitivity in humans. We
previously proposed that DSB repair dependent upon ATM signalling might
involve complex lesions refractory to NHEJ due to additional damage requiring
processing (Riballo et al., 2004). This model was proposed, in part, because loss
of ATM and Artemis, an end-processing nuclease, cause identical and epistatic
DSB repair defects. Additionally, an increased reliance upon ATM/Artemis was
observed for the repair of more complex lesions generated by densely ionizing α-
particles, while they were dispensable for repairing etoposide-induced lesions,
which were predicted not to be associated with damaged bases/sugars (Riballo
et al., 2004). In this model, ATM signalling was thought to regulate Artemis by
phosphorylation. However since Artemis activation has been shown to involve
DNA-PK autophosphorylation and not phosphorylation by ATM, the precise role
of ATM in DSB repair has yet to be defined (Goodarzi et al., 2006).
Here, we consider an alternative model where those lesions that persist in
the absence of ATM signalling are refractory to repair due to the nature of the
surrounding chromatin. There is mounting evidence that dynamic changes to
chromatin play an important role in the DSB response. Laser-induced damage
results in chromatin relaxation in the area surrounding the DSB (Kruhlak et al.,
2006). Studies in yeast have shown that Mre11/Rad50/Xrs2-dependent histone
eviction is required for the proper recruitment of DSB repair factors (Tsukuda et
al., 2005).Chromatin modifierssuchasNuA4 histone-acetyltransferase
complexes, recruited by histone H2A phosphorylation, are also needed for
efficient DSB repair (Downs et al., 2004). Moreover, both yeast and mammalian
systems indicate that higher-order chromatin serves as a barrier to the expansion
of ATM-dependent H2AX phosphorylation (Kim et al., 2007); (Cowell et al., 2007)
and that reduced chromatin compaction (by linker histone H1 depletion)
enhances ATM signalling, checkpoint hypersensitivity and radioresistance
(Murga et al., 2007).
Direct connections between chromatin alteration and ATM were made
when the transcriptional co-repressor KAP-1 (KRAB-Associated Protein 1) was
identified as an ATM substrate, being robustly phosphorylated at S824 and
causing transient chromatin relaxation (Ziv et al., 2006). KAP-1 (also called
TIF1β, TRIM28 or KRIP-1) is an abundant nuclear protein that binds to KRAB
(Krüppel-Associated Box) domains within sequence-specific transcriptional
repressors to trigger heterochromatin formation via interactions with proteins
such as Heterochromatin Protein 1 (HP1), Histone Deacetylases (HDACs), SET-
domain histone methyltransferases and ATP-dependent chromatin remodellers
(reviewed in Craig, 2005).
Altogether, alterations to chromatin are emerging as an increasingly
important part of ATM-signalling. Importantly, the impact of complex higher-order
chromatin (particularly heterochromatin) on DSB repair has not been previously
considered and, moreover, it should be appreciated that many previous studies
were carried out in yeast, which has minimal heterochromatin compared to
Here, we provide mechanistic insight into the role of ATM in DSB repair.
We show that ATM-dependent DSB repair does not correlate with increased
damage complexity but rather overlaps with heterochromatinization. We
demonstrate that ATM signalling facilitates DSB repair by affecting the
transcriptional silencing complex formed by the heterochromatin-building factors
KAP-1, HP1 and/or HDAC1/2. Moreover, we provide evidence that the means by
which ATM signalling enables repair within compacted regions is to transiently
loosen interactions between KAP-1 and the heterochromatin superstructure.
These findings represent the first report of differential DSB repair within
heterochromatin versus euchromatin and demonstrate that an increased reliance
upon ATM for repair correlates with increased chromatin complexity.
A maximum of ~25% of DSBs require ATM-signalling for repair
The ATM DSB repair fraction was first characterized as 10-15% of γ-ray induced
DSBs (Riballo et al., 2004). Notably, densely ionizing α-particles increased ATM
DSB repair to 20-25%. Based on these findings, we suggested that increased
lesion complexity at the DSB might account for an increased reliance upon ATM.
To substantiate this, we examined whether reagents that induce DSBs with
obligatory, multiply damaged termini increase ATM-dependent repair beyond
25%. To monitor DSB repair, we enumerated the rate of γH2AX foci loss by
immunofluorescence. We and others have utilized this technique to measure
DSB repair, the primary advantage being that lower, more clinically relevant
doses of radiation or radiomimetic drugs can be used (compared to pulsed field
gel electrophoresis) and that cells with discrete characteristics (e.g. protein
expression) can be identified.
Neocarzinostatin (NCS) is a radiomimetic reagent that results in defined
DSBs possessing phosphoglycolate or 3’-phosphate termini refractory to ligation
(Dedon and Goldberg, 1992). Significantly, NCS-treatment induces a similar level
of ATM-dependent DSB repair to γ-rays (Figure 1AB; Suppl. Figure 1).
Comparable results were found using the related compound calicheamicin (CLM)
(data not shown). Complex DSBs can be considered to be ‘dirty’, having
damaged bases or sugars at the termini (such as those induced by NCS or CLM)
or multiply-damaged, with additional damages in close proximity. Previously, we
used α-particles to induce multiply-damaged DSBs (Riballo et al., 2004).
However definitive interpretation of this data was complicated by the induction of
many DSBs in extreme proximity, rendering the resolution of γH2AX analysis
difficult. Carbon-K characteristic X-rays (CKX) induce multiply-damaged but
otherwise isolated (and more easily resolved) DSBs arising from ~7 ionisations
within 2 nm2. Notably, while CKX-induced DSBs were repaired with comparatively
slower kinetics to γ-rays or NCS, ATM-dependent repair still accounted for no
more than ~20-25% of overall repair (Figure 1C). Taken together, the percentage
of DSBs that require ATM for repair only weakly correlates with the predicted
complexity of damage, and fails to exceed ~25% even when the damage is
homogenous and/or highly complex.
DSBs dependent upon ATM for repair are associated with heterochromatin
While considering the data described above, we were struck by the fact that the
percentage of the genome consisting of heterochromatin is also in the range of
10-25% (reviewed in(Yunis and Yasmineh, 1971); (Miklos and John, 1979).
Therefore, we addressed whether any correlation between ATM-dependent DSB
repair and heterochromatin could be observed. Murine acrocentric chromosomes
possess clustered regions of satellite DNA identifiable by dense DAPI staining in
NIH3T3 cells (Figure 2A,B). These well-characterized, cytologically distinct
structures are known as ‘chromocenters’ and correspond to pericentric and
centromeric heterochromatin (Guenatri et al., 2004). To establish this under our
conditions, we stained for heterochromatin and euchromatin factors by
immunofluorescence. Consistent with previous work, the densely staining DAPI
regions were enriched for K9 tri-methylated histone H3 and the centromeric
protein CENP-A, markers of heterochromatinization (Guenatri et al., 2004;
Schotta et al., 2004), while the intervening euchromatin was unstained (Figure
2AB). KAP-1 and the three HP1 isoforms (αβγ) stained more strongly within
heterochromatin compared to euchromatin. The transcription factor E2F1 and the
chromatin-distorting HMGB1 protein densely stained within nucleoli but were
absent from heterochromatin, consistent with the silence of genes within these
We next examined whether IR-induced foci (IRIF) that persist in irradiated
cells treated with and ATM inhibitor (ATMi) overlap or juxtapose with
heterochromatin. We utilised contact-inhibited, G1 phase NIH3T3 cells to avoid
complications fromcell division orreplication onnuclear morphology.
Remarkably, most γH2AX or 53BP1 foci persisting in the presence of ATMi were
juxtaposed with heterochromatic regions (Figure 2B, Suppl. Figure 2A-C).
Identical results were obtained using ATM-deficient MEFs (Suppl. Figure 2D),
confirming our data using the ATMi. γH2AX foci were rarely observed to fall
within a chromocenter, instead clustering around the periphery suggesting that
γH2AX expansion and intact heterochromatin may be mutually exclusive. To
explore this further, high resolution, deconvolved images were analyzed for
γH2AX foci / DAPI chromocenter overlap (Figure 2C, i-iv and Suppl. Figure
3A,B). Strikingly, a limited but significant degree of overlap was observed in cells
manifesting the ATM-dependent DSB repair defect. We confirmed this by 3D
modelling, with the majority of γH2AX foci remaining in an ATM-inhibited cell
physically encroaching upon a chromocenter in one or more planes of view
(Figure 2C, v-x). To quantify these findings, we enumerated the total number of
γH2AX foci (Figure 2D) and those observed to juxtapose with heterochromatin
(Figure 2E)(theseanalyses wereperformed blindly). Bysubtracting
heterochromatic γH2AX foci from the total, we estimated the repair kinetics for
euchromatin versus heterochromatin (Figure 2F). Strikingly, DSB repair within
heterochromatin was roughly two-fold slower than repair within regions of
euchromatin. Moreover, ATMi profoundly inhibited the rate of heterochromatic
DSB repair, while having little effect on repair within euchromatin. As a control,
we addressed whether we could detect the anticipated difference in the
percentage of heterochromatin-associated DSBs at early times following low
doses of IR (i.e. stochastic heterochromatin association) compared to later times
after high-dose IR with ATMi (i.e. enriched heterochromatin association) (Suppl.
Figure 4A,B). 30 min after 0.25-2.0 Gy IR, the percent of heterochromatin-
associated DSBs was ~25%, consistent with stochastic induction. By contrast,
the percentage of heterochromatin-associated DSBs at 24 hours post 1-3 Gy IR
with ATMi was >70%.
As an alternative approach, we next measured the persistence of
chromosome breaks within heterochromatinized human chromosomes. The
‘surplus’ female X-chromosome is silenced by heterochromatinization early in
development (Yunis and Yasmineh, 1971); (Miklos and John, 1979). Using
XXXXY cells from a patient with Klinefelter syndrome, in which three of the four
X-chromosomes are fully heterochromatinized (Chapelle et al., 1975), we scored
the number of chromosome breaks persisting following exposure to IR ± ATMi.
Cells were irradiated while in G0/G1-phase, incubated for 24 hrs (to allow repair),
sub-cultured, allowed to enter G2 phase and treated with calyculin-A to promote
chromosome condensation for scoring chromosome breaks. Chromosomes 7
and 8, of comparable size to the X chromosome, were scored as transcriptionally
normal controls. As expected, ATM inhibition resulted in an increased number of
chromosome breaks compared to the DMSO-treated control. Strikingly, ~2-fold
more breaks per metaphase persisted in the X-chromosome compared to
chromosomes 7 and 8 (Figure 3A).
We next determined whether DSB repair in cells with perturbed
heterochromatin had a diminished reliance upon ATM-signalling. Knockout of the
Suv39H1 and Suv39H2 histone methyltransferases in MEFs leads to a reduction
in histone H3 trimethylation, altered DNA methylation and HP1 localization within
Suv39H1/2+/+and Suv39H1/2-/-MEFs were treated ± ATMi, irradiated, harvested
heterochromatin (Lehnertz etal.,2003).Contact-inhibited
and processed for H2AX analysis as indicated (Figure 3B). Although DSB repair
was mildly slower in the Suv39H1/2-/-MEFs compared to WT MEFs, the addition
of ATMi had a significantly smaller impact upon the Suv39H1/2-/-MEFs, with only
half the number of foci remaining at 24-48hrs compared to the ATMi-treated WT
control. Similar results were obtained using fibroblasts from patients with
Hutchinson-GilfordProgeria Syndrome (HGPS),a diseasetypified by
progressive heterochromatic dysfunction (Shumaker et al., 2006). ATMi-treated
HGPS patient cells also had fewer γH2AX foci remaining compared to ATMi-
treated control cells (Figure 3C). Importantly, these data substantiate the link
between heterochromatinization and ATM-dependent DSB repair.
Finally, we utilized a method not reliant upon microscopy to observe the
correlation between heterochromatin and ATM-dependent DSB repair. Using
γH2AX antibodies, we immunoprecipitated (IP) polynucleosomes (repeat length
from 1 to ~10) proximal to persistent DSBs and blotted for heterochromatic
markers (histone trimethylation) and euchromatic markers (histone acetylation)
(Figure 3D-F and Suppl. Figure 5). In the absence of ATM, γH2AX foci are less
intense (Stiff et al., 2004); this reduced ‘quality’ (but not quantity) of γH2AX foci
manifests as a reduction in epitope for IP. To control for this, ATMi was removed
0.5 hour before harvesting to enable γH2AX intensity to return to normal but not
enough time for repair (data not shown); this proved to be critical to IP similar
amounts of nucleosomes ± ATMi. We first determined conditions whereby similar
γH2AX foci numbers were obtained using ATMi, high IR and late times (= ATM-
dependent DSB repair fraction) versus DMSO, low IR and early times (=
stochastic DSBs) (Figure 3D). Using these conditions, equivalent levels of
immunoprecipitated γH2AX in ATM-inhibited cells were associated with ~3X
more histone H3 K9 trimethylation and ~6X less histone H3 K9 acetylation than
cells with stochastically incurred DSBs (Figure 3E,F). This confirms our
microscopy data and strongly suggests that ATM signalling is important for DSB
repair within heterochromatin.
KAP-1 knockdown alleviates the DSB repair defect of ATM-inhibited cells
ATM phosphorylatesthe KAP-1 co-repressor,a core componentof
heterochromatin (Ziv et al., 2006). This raised the possibility that ATM signalling
might modify compacted chromatin nearby DSBs by phosphorylating KAP-1,
modulating its interaction with silencing factors and thereby facilitating repair. To
test this, we examined whether ATM-dependent repair was affected by KAP-1
knockdown using the shRNA system employed previously (Ziv et al., 2006).
Knockdown was verified by staining for KAP-1 and γH2AX, allowing selective
analysis of cells without KAP-1 expression (Figure 4A,B). GFP shRNA vectors
were used as controls. ATMi treatment resulted in the expected DNA repair
defect in the GFP shRNA control (Figure 4C) or mock-transfected cells (data not
shown). Qualitatively, γH2AX foci in ATMi treated cells were less intense in
control cells. Strikingly, ATMi-treated cells with KAP-1 knockdown completed
DSB repair with wildtype kinetics. This suggests that KAP-1 disruption removes
the barrier to repair that ATM normally alleviates. Of interest, a qualitative
increase in the size (but not number) of γH2AX foci in KAP-1 knockdown cells
was observed, in agreement with studies demonstrating that chromatin
compaction is refractory to γH2AX expansion (Kim et al., 2007).
To further the analysis of the interplay between KAP-1 and ATM, we next
re-introduced KAP-1 and ATM phosphorylation site mutants of KAP-1 (S824A or
S824D) into cells with ablated endogenous KAP-1. 1BRneo cells were
transfected with KAP-1 siRNA and pEGFP expression vectors encoding siRNA-
resistant GFP-tagged WT, S824A or S824D KAP-1 cDNA (Figure 4E). Cells were
then treated ± ATMi, irradiated and processed for 53BP1 foci analysis. Foci were
enumerated in cells with verified knockdown and/or GFP signal (Suppl. Figure 6).
As before, KAP-1 knockdown relieved the need for ATM-signalling in DSB repair
(Figure 4F). Re-expression of GFP-KAP-1WTin cells with no endogenous KAP-1
expression of GFP-KAP-1S824A(ATM phospho-mutant) induced a DSB repair
defect with or without ATMi-treatment while cells expressing GFP-KAP-1S824D
ATM-dependentrepair, confirmingspecificity. Dramatically,the
(ATM phospho-mimic) had normal repair kinetics even in the presence of ATMi.
These striking data strongly suggesting that ATM-signalling facilitates DSB repair
by phosphorylating KAP-1.
We also ablated KAP-1 using siRNA of a distinct sequence to that
described above. KAP-1 specific knockdown was efficiently achieved 48 hours
post transfection (Suppl. Figure 7). Similar results to those described in Figure 4
were obtained, where KAP-1 knockdown enabled normal DSB repair in the
presence of ATMi (Figure 5A). We conclude that the presence of KAP-1 is
inhibitory to DSB repair in the absence of ATM-signalling. To substantiate this
further, we examined whether other heterochromatic factors influenced ATM-
dependent DSB repair.
HP1 or HDAC1/2 knockdown also bypasses the requirement for ATM in
A well-characterized binding partner of KAP-1 is HP1, an adapter protein
associated with constitutive and facultative heterochromatin (Craig, 2005).
Localized to sites of transcriptional repression by KAP-1, HP1 interacts with
methyltransferases to promote histone H3 K9 tri-methylation and undergoes
inter-nucleosomal oligomerization to promote chromatin compaction by cross-
linking (Yamada et al., 1999). HP1 ablation was achieved using siRNA to all
three isoforms of HP1 (α/β/γ). Knockdown was verified by immunofluorescence
(Suppl. Figure 7) and γH2AX foci were scored specifically in non-expressing
cells. Significantly, HP1 loss enabled normal repair kinetics with ATMi (Figure
5B). The effect of HP1 knockdown was indistinguishable to KAP-1, indicating that
they likely have an over-lapping impact upon repair.
Histone deactylases (HDACs) are another important component of
heterochromatin (Narlikar et al., 2002). Opposing histone acetyltransferases and
often in complex with ATP-dependent chromatin-remodellers, HDACs are
targeted to newly deposited histones to remove transcriptionally activating acetyl-
groups within heterochromatin. HDAC1 and HDAC2 are components of NuRD, a
chromatin-remodeling complex of which the Mi-2α subunit interacts with KAP-1 to
participate transcriptional repression (Schultz et al., 2001). Since both HDAC1
and HDAC2 are found within heterochromatic DNA remodeling complexes(Craig,
2005) and likely have redundant activity, we ablated their expression individually
and together. While knockdown of either HDAC1 or HDAC2 alone had no
significant effect on ATM-dependent DSB repair (Figure 5C), combined
knockdown together with ATMi resulted in wildtype repair kinetics (Figure 5D).
These data suggest that perturbation of heterochromatin by loss of the co-
repressor (KAP-1), adaptor (HP-1) or HDAC components of its foundation
sufficiently relaxes otherwise inaccessible regions of chromatin to enable DSB
repair in the absence of active ATM.
KAP-1 knockdown alleviates the DSB repair defect of ATM-deficient MEFs
To confirm our findings obtained with ATMi, we ablated KAP-1 expression in
ATM-/-MEFs. As expected, KAP-1 knockdown alleviated the DSB repair defect
observed in ATM-/-cells (Figure 5E). Identical results were obtained using
NIH3T3 cells treated with ATMi (data not shown). Transient KAP-1 knockdown
did not visibly affect the presence of DAPI-chromocenters in NIH3T3 cells (Suppl.
Figure 8A,B). However, although chromocenters were apparently intact after
KAP-1 siRNA, the level of histone H3 K9 trimethylation within them was altered,
as previously observed after Suv39H1/2 knockout (Lehnertz et al., 2003). No
alteration in HP-1 localization to chromocenters was observed following KAP-1
knockdown (data not shown), in agreement with previous data (Matsuda et al.,
2001). Finally, we confirmed our findings by measuring DSB repair using pulsed-
field gel electrophoresis in WT MEFs with ATM and/or KAP-1 knockdown or AT
MEFs with KAP-1 knockdown. As expected, KAP-1 knockdown reversed the
DNA repair defect associated with ATM deficiency (Figure 5F,G; Suppl. Figure
KAP-1 association with nuclease-resistant chromatin changes after IR
Previously, KAP1 phosphorylation was shown to correlate with global chromatin
relaxation as measured by micrococcal nuclease (MNase) digestion of genomic
DNA, although the underlying mechanism was unclear (Ziv et al., 2006). Here,
we examined whether irradiation and ATM-signalling might impact upon the
chromatin retention of KAP-1. The preparation of non-denatured, soluble
chromatin requires nuclease digestion and high salt extraction (Yoda and Ando,
2004). Increased nucleosome compaction, as caused by heterochromatinization,
is more resistant to solublization by this technique (Yoda and Ando, 2004). We,
therefore, used a strategy of incrementally increasing the salt and nuclease
treatment to separate cell extracts into fractions enriched for highly soluble
proteins, open chromatin proteins or compacted chromatin proteins (Figure 6A,
Suppl. Figure 5). As expected, KAP-1, HP1 and HDAC1 were abundant in
fractions most resistant to nuclease digestion. Proteins associated with open
chromatin (HMGB1) were absent from this fraction. Strikingly, IR caused a
reduction of KAP-1 in the most nuclease-resistant fraction (Figure 6A,B). This
depletion progressively increased between 2 and 20 Gy IR, beyond which there
was no further reduction. A small percentage of KAP-1 remained even following
80 Gy. The reduced association of KAP-1 was reversed within several hours,
mirroring the loss of γH2AX and indicative of a dynamic equilibrium with repair.
Most importantly,KAP-1 releasedidnot occurwith ATMi-treatment,
demonstrating its dependence upon ATM-signalling (Figure 6C). Under the
conditions of this experiment, HP1 and HDAC1 did not significantly disperse from
the nuclease-resistant fraction after IR.
Here, we focused on addressing the role of ATM signalling in DSB repair. We
demonstrate for the first time that DSBs that persist in the absence of ATM-
signalling localize to heterochromatin, that heterochromatinized chromosomes
accumulate increased breaks in the absence of ATM activity and that cells with
constitutively perturbed heterochromatin have a smaller reliance upon ATM for
DSB repair. Knockdown or mutation of heterochromatin building factors (KAP-1,
HP1, HDAC1/2, Suv39H1/2) bypasses or reduces the requirement for ATM in
DSB repair, suggesting that an important end-point of ATM-signalling is to
regulate the heterochromatic super-structure to facilitate repair. Importantly, cells
expressing KAP-1 that cannot be phosphorylated by ATM display a constitutive
DSB repair defect, while cells mimicking phosphorylated KAP-1 bypass ATM-
dependent repair. Moreover, following DSB induction the association of
endogenous KAP-1 with nuclease-resistant chromatin is reduced in an ATM-
dependent manner. We propose that the heterochromatic barrier to transcription
blocks DSB repair and that ATM signaling functions to temporarily remove this
A model for the role of ATM-signalling in DSB repair
Based upon these data and previous work (Lavin et al., 2005; Ziv et al., 2006),
we propose the following model. DSB formation triggers ATM activation. In
euchromatin, ATM-signalling perturbs the local chromatin architecture that, while
important for optimal signalling, is dispensable for repair since NHEJ factors can
freely access or manipulate the DSB. Hence, >75% of DSBs in G0/G1 are
repaired in an ATM-independent manner. By contrast, in regions where
nucleosome flexibility is constrained by heterochromatic factors (KAP-1, HP1,
HDACs, etc.), repair proteins are unable to adequately access or manipulate the
DSB and require ATM to phosphorylate KAP-1, diminish KAP-1 interactions with
heterochromatin and thereby provide sufficient elasticity to facilitate repair. In the
absence of ATM-signalling, heterochromatic repair stalls and the lesion persists.
Interestingly, lower eukaryotes, which have minimal heterochromatin,
have a diminished role for ATM-signalling in their DSB response (Morrow et al.,
1995). By contrast, ATM-signalling is of major importance to the higher
eukaryotic DSB response (Lavin et al., 2005), where a larger genome and
complex developmental program has resulted in more constitutive and facultative
heterochromatin. We suggest that the increased prominence of ATM-signalling in
the DSB response (compared to TEL1-signalling in yeast) correlates with the
increasing complexity of chromatin architecture observed through eukaryotic
The discovery of a connection between ATM and heterochromatic DSB
repair prompts a reassessment of previous models. We originally postulated that
the ATM-dependent DSB repair defect might represent lesions refractory to
repair due to increased complexity. Results using heavily ionizing α-particles that
increased ATM-dependent repair compared to ‘clean-DSB‘-inducers (etoposide)
that were ATM-independent, supported this conclusion (Riballo et al., 2004).
Further analysis using radiation types/qualities that induce obligatory multiply
damaged termini argues against this hypothesis. Indeed, under conditions where
DSBs have homogenous damage complexity (requiring processing), the
percentage of DSBs requiring ATM-signalling for repair remains at 10-25% and
our analysis here indicates that the ‘complexity’ necessitating ATM-signalling
relates to the nature of the surrounding chromatin rather than damage
complexity. We are cautious, however, in eliminating the possibility that lesion
complexity plays some role, since heterochromatin interferes with the processing
phase of NHEJ rather than end-joining per se. Future work will be required to
distinguish these possibilities and to address the role of Artemis. Indeed, the
absolute requirement for Artemis and DNA-PK in the repair of ATM-dependent
lesions (Riballo et al., 2004) may suggest that it is the processing of complex
DSBs that stalls within rigid chromatin. We are currently pursuing these studies.
Our findings are also relevant to recent reports suggesting that ATM-
signalling enhances DSB repair within nucleoli (Berkovich et al., 2007); (Kruhlak
et al., 2007). Nucleoli may be as problematic for repair as heterochromatin, due
to the high density of transcription complexes that restrict the accessibility and/or
flexibility of rDNA. Kruhlak and colleagues (Kruhlak et al., 2007) demonstrated
that DSB formation triggers the inhibition of RNA Pol I in the nucleolus in a
process requiring ATM, MDC1 and NBS1. Berkovich et al., 2007 found that
enzymatically induced DSBs within nucleoli are repaired in an ATM/NBS1-
dependent manner over a 16-hour period. Since the heterochromatinization of
rDNAis an established mechanism for regulating ribosome production
(Akhmanova et al., 2000; Carmo-Fonseca et al., 2000) it is possible to reconcile
the idea of ATM-dependent nucleolar repair with our current findings.
Our results also add to the interpretation of previous work using MNase to
produce polynucleosomes from genomic DNA, the largest of which decreased in
size in an ATM/KAP-1 phosphorylation dependent manner after DSB induction
(Ziv et al., 2006). Although proposed to represent global chromatin relaxation, we
now postulate that the larger polynucleosomes might represent nuclease-
resistant chromatin enriched for heterochromatin factors. Thus, the DSB-induced
reduction in maximum nucleosome size following DSB formation could represent
the loosening of KAP-1 dependent heterochromatin, allowing enhanced nuclease
access, rather than a global relaxation per se.
The ATM-dependent KAP-1 phosphorylation site (S824Q) resides within
the C-terminus encompassing the HP1 interaction region, the Plant-Homeo-
Domain and Bromo-Domain, which are important for recruiting heterochromatin
factors (Schultz et al., 2001). The N-terminal RBCC domain (RING finger, B-box,
Coiled-Coiled), although sufficient for interaction with sequence specific KRAB
repressors, cannot trigger transcriptional repression as it does not mediate
heterochromatic factor interactions (Matsuda et al., 2001). Thus, ATM-dependent
C-terminal KAP-1 phosphorylation may only perturb interactions with silencing
factors (HP1, HDACs) without affecting interactions that target KAP-1 to
chromatin. Thus, we hesitate to suggest that IR triggers the release of KAP-1
from heterochromatin in vivo. A more cautious explanation for the data described
here is that IR weakens interactions between KAP-1 and heterochromatin that,
under the conditions of our assay, manifests as in vitro dissociation. This
heterochromatic ‘loosening’ could, however, provide the critical flexibility to
facilitate repair within highly compacted regions without necessitating the
disassembly of the heterochromatic superstructure.
In summary, our findings provide mechanistic insight into how higher order
chromatin structure impacts upon the DNA DSB response influencing not only
signalling, as previously shown, but also repair. Strikingly, we show that ATM is
preferentially required for DSB repair within heterochromatin demonstrating a
new aspect of ATM signalling.
Materials and Methods
1BR3 (WT), HSF3 (WT), 48BR (WT) and AT1BR (ATM-defective) primary
human fibroblasts were cultured as described (Riballo et al., 2004). NIH3T3 cells
and ATM-deficient MEFs (Riballo et al., 2004) were cultured in DMEM with 10%
FCS, L-glutamine, penicillin and streptomycin. 1BRneo (WT) transformed human
fibroblasts were cultured in MEM supplemented as above. AG11513 (HGPS) and
GM00157 (Klienfelter syndrome) primary human fibroblasts were obtained from
Coriell Cell Repositories (Camden, NJ, USA) and cultured in MEM supplemented
Transient knockdown of protein expression
The pRETRO-SUPER shRNA system used for KAP-1 (or GFP control)
knockdown was as described in (Ziv et al., 2006). Knockdown of KAP-1, HP1α,
HP1β, HP1γ, HDAC1 and/or HDAC2 was achieved by either siPORT™NeoFX™
(Ambion) or Metafectene®-Pro (Biontex, Germany) mediated transfection
(according to the manufacturer’s instructions) using 100 pmol of siRNA duplexes
per 2 x 105cells (1BRneo, NIH 3T3, ATM+/+MEFs or ATM-/-MEFs). siRNA
duplexes were Stealth RNAi synthesized by Invitrogen:
KAP-1‘A’ =CAGUGCUGCACUAGCUGUGAGGAUA (human cDNA nt 450-475)
KAP-1‘B’ =GCAUGAACCCCUUGUGCUGUUUUGU (human cDNA nt 928-947)
KAP-1‘C’ =AAGAUGCAGUGAGGAACCAACGUAA (mouse cDNA nt 1297-1322)
HP1α =GGUUAAGGGACAAGUGGAAUAUCUA (human cDNA nt 89-114)
HP1β =AAGGGCAAAGUGGAGUACCUCCUAA (human cDNA nt 113-158)
HP1γ =CAGAGGUCUUGAUCCUGAAAGAAUA (human cDNA nt 203-228)
HDAC1 =UCUUGCGCUCCAUCCGUCCAGAUAA (human cDNA nt 224-249)
HDAC2 =CACCUGGUGUCCAGAUGCAAGCUAU (human cDNA nt 1381-1406)
Knockdown was observed at early as 24 hr post-transfection, but was not
optimal until 48-72 hr. For DSB repair experiments, cells were irradiated 48 hr
post-transfection and harvested up to 24 hr later (i.e. during the window of
Nucleosome solublization and γH2AX IP
5 x 107cells were washed with PBS and once with low salt buffer (LSB: 10
mM HEPES pH 7.4, 25 mM KCl, 10 mM NaCl, 1 mM MgCl2, 0.1 mM EDTA).
Pelleted cells were resuspended in 5X the packed cell volume (PCV) of LSB +
0.1 μM MC-LR and 1X protease inhibitor cocktail (Sigma-Aldrich, UK) and snap-
frozen in liquid nitrogen. Cells were quick-thawed and immediately centrifuged for
10 min at 10000 rpm. The resulting supernatant, containing cytoplasmic and
loosely bound nuclear proteins, was discarded and the pellet resuspended in 2X
the original PCV of nuclease buffer (10 mM HEPES, pH 7.9, 10 mM KCl, 1.0 mM
CaCl2,1.5 mM MgCl2,0.34 M Sucrose, 10% Glycerol, 1 mM DTT, 0.1 % (v/v)
Triton X-100) containing 100 U/mL MNase. Samples were incubated at 37oC for
45 min before addition of an equal volume of solublization buffer (nuclease buffer
+ 2% (v/v) NP-40, 2% (v/v) Triton X-100, 600 mM NaCl). Samples were then
sonicated briefly and centrifuged at 10000 rpm for 10 min. The resulting
supernatant, containing solublized nucleosomes, was incubated with 2 μL of anti-
γH2AX monoclonal antibody overnight at 4oC. Immuno-complexes were pulled
down by adding 15 μL of protein G-sepharose for 30 min at 4oC and washed
three times with wash buffer (nuclease buffer + 1% (v/v) NP-40, 1% (v/v) Triton
X-100, 300 mM NaCl). 20 μL 2X SDS sample buffer was added to each
immunoprecipitate and samples were boiled for 2 min before electrophoresis.
Chromatin Segregation Assay
1 x 107cells were washed with PBS and once with 1 mL LSB. Pelleted
cells were resuspended in 6X the PCV of LSB + 0.1 μM MC-LR and 1X protease
inhibitor cocktail and snap-frozen in liquid nitrogen. Samples were quick-thawed
and immediately centrifuged for 10 min at 10000 rpm (supernatant = S10). The
pellet was gently resuspended (by tapping, but not pipetting to prevent chromatin
decondensation) in a volume (V) of high salt buffer (HSB: 50 mM Tris-HCl, pH
8.0, 5% (v/v) glycerol, 1 mM EDTA, 10 mM MgCl2, 400 mM KCl, 1X protease
inhibitors and 0.1 μM MC-LR) equal to 0.25 V of LSB used to lyse the cells.
Samples were immediately centrifuged for 10 min at 10000 rpm (supernatant =
P10). The pellet was then resuspended in nuclease buffer (same V as HSB)
containing 10 U/mL MNase and incubated at 37oC for 10 min. Samples were
then centrifuged for 5 min at 10000 rpm (supernatant = C1). The pellet was
resuspended in nuclease buffer (V same as used for HSB) containing 100 U/mL
MNase and was incubated at 37oC for 45 min before an equal V of solublization
buffer (nuclease buffer + 2% (v/v) NP-40, 2% (v/v) Triton X-100, 600 mM NaCl)
was added. Samples were vortexed briefly and centrifuged for 5 min at 10000
rpm (supernatant = C2). The remaining pellet was resuspended in solublization
buffer (same V as HSB) and an equal V of denaturing buffer (50 mM Tris pH 6.8,
1% (v/v) SDS, 100 mM DTT, 10% glycerol) before being brief sonication, boiling
for 5 min and centrifuging at 10000 rpm for 5 min (supernatant = C3).
See Supplemantary Materials for:
Nucleosome Size Control, Chromosomal Break Analysis, Pulsed Field Gel
Electrophoresis and Expression of GFP-KAP-1 in KAP-1 knockdown cells
We thank Dr. Thomas Jenuwein (Research Institute of Molecular Pathology,
Austria), Dr. Ola Hammarsten (Sahlgrenska University Hospital, Sweden), Dr.
Stefan Hofreiter (BiontexGmbH, Germany) and Dr Graeme Smith (KuDos
Pharmaceuticals, UK) for generously providing reagents. AAG is supported by a
Post-Doctoral Fellowship grant from the Alberta Heritage Foundation for Medical
Research. The PAJ laboratory is funded by the MRC, the Leukaemia Research
Fund, the Department of Health (UK), the EU (Rad-Risc and DNA Repair) and
the International Association for Cancer Research. The ML laboratory is funded
by the Deutsche Forschungsgemeinschaft, the Bundesministerium für Bildung
und Forschung via the Forschungszentrum Karlsruhe and the Wilhelm Sander-
Akhmanova, A., Verkerk, T., Langeveld, A., Grosveld, F., and Galjart, N. (2000).
Characterisation of transcriptionally active and inactive chromatin domains in
neurons. J Cell Sci 113 Pt 24, 4463-4474.
Berkovich, E., Monnat, R.J., Jr., and Kastan, M.B. (2007). Roles of ATM and
NBS1 in chromatin structure modulation and DNA double-strand break repair.
Nat Cell Biol 9, 683-690.
Carmo-Fonseca, M., Mendes-Soares, L., and Campos, I. (2000). To be or not to
be in the nucleolus. Nat Cell Biol 2, E107-112.
Chapelle, A., Miller, R.C., Greene, A.E., and Coriell, L.L. (1975). A (4;11)
translocation, balanced, 49 XXXXY karyotype. Repository identification no. GM-
157. Cytogenet Cell Genet 15, 127-128.
Cowell, I.G., Sunter, N.J., Singh, P.B., Austin, C.A., Durkacz, B.W., and Tilby,
M.J. (2007). gammaH2AX Foci Form Preferentially in Euchromatin after Ionising-
Radiation. PLoS ONE 2, e1057.
Craig, J.M. (2005). Heterochromatin--many flavours, common themes. Bioessays
Deckbar, D., Birraux, J., Krempler, A., Tchouandong, L., Beucher, A., Walker, S.,
Stiff, T., Jeggo, P.A., and Lobrich, M. (2007). Chromosome breakage after G2
checkpoint release. Journal of Cell Biology 176, 748-755.
Dedon, P.C., and Goldberg, I.H. (1992). Influence of thiol structure on
neocarzinostatin activation and expression of DNA damage. Biochem. 31, 1909-
Downs, J.A., Allard, S., Jobin-Robitaille, O., Javaheri, A., Auger, A., Bouchard,
N., Kron, S.J., Jackson, S.P., and Cote, J. (2004). Binding of chromatin-
modifying activities to phosphorylated histone H2A at DNA damage sites. Mol
Cell 16, 979-990.
Goodarzi, A.A., Yu, Y., Riballo, E., Douglas, P., Walker, S.A., Ye, R., Harer, C.,
Marchetti, C., Morrice, N., Jeggo, P.A., and Lees-Miller, S.P. (2006). DNA-PK
autophosphorylation facilitates Artemis endonuclease activity. Embo J 25, 3880-
Guenatri, M., Bailly, D., Maison, C., and Almouzni, G. (2004). Mouse centric and
pericentric satellite repeats form distinct functional heterochromatin. J Cell Biol
Kim, J.A., Kruhlak, M., Dotiwala, F., Nussenzweig, A., and Haber, J.E. (2007).
Heterochromatin is refractory to gamma-H2AX modification in yeast and
mammals. J Cell Biol 178, 209-218.
Kruhlak, M., Crouch, E.E., Orlov, M., Montano, C., Gorski, S.A., Nussenzweig,
A., Misteli, T., Phair, R.D., and Casellas, R. (2007). The ATM repair pathway
inhibits RNA polymerase I transcription in response to chromosome breaks.
Nature 447, 730-734.
Kruhlak, M.J., Celeste, A., and Nussenzweig, A. (2006). Spatio-temporal
dynamics of chromatin containing DNA breaks. Cell Cycle 5, 1910-1912.
Lavin, M.F., Birrell, G., Chen, P., Kozlov, S., Scott, S., and Gueven, N. (2005).
ATM signaling and genomic stability in response to DNA damage. Mutat Res
Lehnertz, B., Ueda, Y., Derijck, A.A., Braunschweig, U., Perez-Burgos, L.,
Kubicek, S., Chen, T., Li, E., Jenuwein, T., and Peters, A.H. (2003). Suv39h-
mediated histone H3 lysine 9 methylation directs DNA methylation to major
satellite repeats at pericentric heterochromatin. Curr Biol 13, 1192-1200.
Matsuda, E., Agata, Y., Sugai, M., Katakai, T., Gonda, H., and Shimizu, A.
(2001). Targeting of Kruppel-associated box-containing zinc finger proteins to
centromeric heterochromatin. Implication for the gene silencing mechanisms. J
Biol Chem 276, 14222-14229.
Miklos, G.L., and John, B. (1979). Heterochromatin and satellite DNA in man:
properties and prospects. Am J Hum Genet 31, 264-280.
Morrow, D.M., Tagle, D.A., Shiloh, Y., Collins, F.S., and Hieter, P. (1995). TEL1,
anS. cerevisiaeHomologue ofthe
Telangiectasia, Is Functionally Related to the Yeast Checkpoint Gene MEC1.
Cell 82, 831-840.
Murga, M., Jaco, I., Fan, Y., Soria, R., Martinez-Pastor, B., Cuadrado, M., Yang,
S.M., Blasco, M.A., Skoultchi, A.I., and Fernandez-Capetillo, O. (2007). Global
chromatin compaction limits the strength of the DNA damage response. J Cell
Biol 178, 1101-1108.
Narlikar, G.J., Fan, H.Y., and Kingston, R.E. (2002). Cooperation between
complexes that regulate chromatin structure and transcription. Cell 108, 475-487.
Riballo, E., Kuhne, M., Rief, N., Doherty, A., Smith, G.C., Recio, M.J., Reis, C.,
Dahm, K., Fricke, A., Krempler, A., et al. (2004). A pathway of double-strand
break rejoining dependent upon ATM, Artemis, and proteins locating to gamma-
H2AX foci. Mol Cell 16, 715-724.
Schotta, G., Lachner, M., Sarma, K., Ebert, A., Sengupta, R., Reuter, G.,
Reinberg, D., and Jenuwein, T. (2004). A silencing pathway to induce H3-K9 and
H4-K20 trimethylation at constitutive heterochromatin. Genes Dev 18, 1251-
Schultz, D.C., Friedman, J.R., and Rauscher, F.J., 3rd (2001). Targeting histone
bromodomains of KAP-1 form a cooperative unit that recruits a novel isoform of
the Mi-2alpha subunit of NuRD. Genes Dev 15, 428-443.
Shumaker, D.K., Dechat, T., Kohlmaier, A., Adam, S.A., Bozovsky, M.R., Erdos,
M.R., Eriksson, M., Goldman, A.E., Khuon, S., Collins, F.S., et al. (2006). Mutant
nuclear lamin A leads to progressive alterations of epigenetic control in
premature aging. Proc Natl Acad Sci U S A 103, 8703-8708.
Stiff, T., O'Driscoll, M., Rief, N., Iwabuchi, K., Lobrich, M., and Jeggo, P.A.
(2004). ATM and DNA-PK function redundantly to phosphorylate H2AX following
exposure to ioninsing radiation. Cancer Res. 64, 2390-2396.
Tsukuda, T., Fleming, A.B., Nickoloff, J.A., and Osley, M.A. (2005). Chromatin
remodelling at a DNA double-strand break site in Saccharomyces cerevisiae.
Nature 438, 379-383.
Yamada, T., Fukuda, R., Himeno, M., and Sugimoto, K. (1999). Functional
domain structure of human heterochromatin protein HP1(Hsalpha): involvement
of internal DNA-binding and C-terminal self-association domains in the formation
of discrete dots in interphase nuclei. J Biochem (Tokyo) 125, 832-837.
Yoda, K., and Ando, S. (2004). Immunological analysis and purification of
centromere complex. Methods Enzymol 375, 270-277.
Yunis, J.J., and Yasmineh, W.G. (1971). Heterochromatin, satellite DNA, and cell
function. Structural DNA of eucaryotes may support and protect genes and aid in
speciation. Science 174, 1200-1209.
Ziv, Y., Bielopolski, D., Galanty, Y., Lukas, C., Taya, Y., Schultz, D.C., Lukas, J.,
Bekker-Jensen, S., Bartek, J., and Shiloh, Y. (2006). Chromatin relaxation in
response to DNA double-strand breaks is modulated by a novel ATM- and KAP-1
dependent pathway. Nat Cell Biol 8, 870-876.
Human Gene Mutatedin Ataxia
Figure 1: A maximum of ~25% of DSBs require ATM-signalling for repair.
Panel A: 1BR3 (WT) and AT1BR (ATM-/-) primary human fibroblasts were
irradiated (or not) with 2 Gy (γ-ray) IR. Panel B: 1BR3 (WT) and AT1BR (ATM-/-)
primary human fibroblasts were treated with 50 ng/mL NCS. Panel C: HSF3 (WT)
or AT1BR (ATM-/-) primary human fibroblasts were irradiated (or not) with 1 Gy
of Carbon-K X-rays. Here and in all subsequent figures where indicated, cells
were fixed at the indicated times, immunostained for γH2AX and foci were
enumerated for >30 cells per condition, blind counting. The asterisk in 1C
indicates that we were unable to count the 0.25 hr time point in the A-T cell due
to delayed foci formation. All results here and subsequently represent the mean
and SD of three experiments.
Figure 2: γH2AX foci remaining in ATMi-treated cells are associated with
regions of heterochromatin. Panel A: Untreated NIH 3T3 cells were fixed and
stained for tri-methylated K9 of histone H3, CENP-A, KAP-1, HP1αβγ, HMGB1,
E2F1 or Ku70/80 as indicated (red) before being counterstained with DAPI
(blue). Panel B: DMSO or 10 μM KU55933 ATM inhibitor (ATMi) was added to
confluent, stationary-phase NIH 3T3 cells and, 0.5 hr later, cells were irradiated
with 2 Gy IR and harvested 24 hr post-IR. Cells were fixed and stained for H2AX
(green), tri-methylated K9 of histone H3 (red) and DAPI (blue). All images shown
are representative of the total cell population. Panel C: ATMi-treated NIH3T3
cells were irradiated with 3 Gy IR and harvested 24 hours later. Cells were fixed
and stained for H2AX (green) and DAPI (blue). High resolution Z-stacks of
deconvolved images were acquired for a representative cell demonstrating the
ATM-dependent repair defect (i-iv). Using softWoRx®, γH2AX foci (ii) and
intensely staining DAPI chromocenters (iii) were isolated and the regions of
overlap between them was assessed (iv, colored peach in ii-iv). Using the
regions isolated in ii-iv, 3D models of γH2AX foci (green) and heterochromatic
chromocenters (blue) were generated (v-x). (vi-x) are rotations of the image
shown in (v). Panel D: NIH3T3 cells were treated as in (B) and harvested at the
indicated time points. Total γH2AX foci numbers were scored. Panel E: γH2AX
foci juxtaposed with the heterochromatic regions (assessed by DAPI staining)
were also scored. Panel F: Using the data generated in C & D, the % remaining
γH2AX foci associated with euchromatin (EC, blue circles) or heterochromatin
(HC, red squares), in the presence of DMSO (filled circles/squares) or ATMi
(open circles/squares) was estimated.
Figure 3: DNA double-strand breaks dependent upon ATM activity for
repair are associated with regions of heterochromatin. Panel A: Confluent
GM00157 (Klinefelter syndrome) cells were treated ±ATMi, irradiated with 2 Gy
IR, incubated 24 hr, subcultured (±ATMi) and harvested after an additional 32 hr
(when the maximum number of G2 phase cells was obtained as determined by
flow cytometry). 50 ng/mL calyculin A was added 0.5 hr prior to harvest to induce
premature chromosome condensation of G2 cells. Fixation and preparation of
chromosome spreads was as described previously (Deckbar et al., 2007). Slides
were processed for FISH against chromosomes 7, 8 and X, and the numbers of
chromosomes breaks per metaphase were scored. Panel B: Stationary-phase
Suv39H1/2 double knockout or WT MEFs were treated ± ATMi, irradiated with 3
Gy and harvested as indicated. Cells were fixed, stained for γH2AX and
enumerated. Panel C: 48BR (WT) and AG11513 (HGPS) primary human
fibroblasts were treated as in (B). Panel D: NIH 3T3 cells were treated ±ATMi,
irradiated and harvested as indicated. Where ATMi was used, the drug was
removed 0.5 hr before harvesting. Cells were lysed and nucleosomes solublized
as described in Methods. 60 μg of soluble nucleosome extract was blotted for
γH2AX and tri-methylated K9 of histone H3 (TriMe K9 H3). A Ponceau-S stain of
visible histones is shown. Duplicate samples were processed for γH2AX
immunofluorescence and the numbers of γH2AX foci were scored. Panel E:
NIH3T3 cells were treated as in (D). γH2AX was immunopreciptiated from 1 mg
of soluble nucleosome extract, washed with solublization buffer and blotted for
γH2AX, TriMe K9 H3 and Acetyl K9 H3. Panel F: The immunoblot data in (E) was
quantified and expressed as % of the maximal TriMe K9 H3 signal divided by %
of maximal γH2AX signal.
Figure 4: Knockdown of KAP1 by shRNA alleviates the DSB repair defect of
1BRneo cells treated with ATM inhibitors. Panel A: 1BRneo cells were
transfected pRETRO-super vector encoding GFP or KAP-1 shRNA. Cells were
fixed and stained for KAP-1 (red) and DAPI (blue). Panels B,C: 24-48 hr post
transfection (as in (A)), cells were treated ± ATMi, irradiated with 3 Gy IR,
harvested and stained for KAP-1 (red), γH2AX (green) and DAPI (blue). In Panel
B, a representative image of an ATMi-treated, irradiated cell 24 hrs post IR
stained for KAP-1 (red), γH2AX (green) and DAPI (blue) are shown. A cell with
KAP-1 knockdown is shown adjacent to an unsuccessfully transfected cell. Panel
C: Representative images of γH2AX and DAPI overlays. Panel D: γH2AX foci
numbers in KAP-1 knockdown cells were quantified. Panel E: 1BRneo cells were
transfected with siRNA duplexes to human KAP-1 (KAP-1 ‘B’, identical sequence
to system used in (A)) or scrambled siRNA (mock). 48 hr later, cells were re-
transfected with siRNA with or without pEGFP vectors encoding WT, S824A or
S824D human KAP-1 siRNA-resistant cDNA. 24 hr later, cells were fixed and
stained for KAP-1 (red) and DAPI (blue). GFP expression = green. Panel F:
1BRneo cells were transfected as in (E) and treated as in (B,C). Cells were fixed
and stained with 53BP1 and DAPI. Numbers of 53BP1 foci were enumerated.
Figure 5: Knockdown of KAP-1, HP1 or HDAC1&2 alleviates the DSB repair
defect of ATM inhibited cells. Panel A: 1BRneo cells were transfected with
siRNA duplexes targeted to human KAP-1 (KAP-1 ‘A’ siRNA) or scrambled
siRNA (mock). 48 hr post transfection, cells were treated ± ATMi, irradiated with
3 Gy IR, harvested and stained for KAP-1 and γH2AX. γH2AX foci were
enumerated in cells with verified knockdown. Panel B: 1BRneo cells were
transfected with scrambled siRNA, combined siRNA duplexes targeted to HP1α,
HP1β and HP1γ (= HP1 siRNA) or HP1 and KAP-1 siRNA as in (A). Cells were
treated as in (A) and stained for HP-1 and γH2AX. γH2AX foci were enumerated
as in (A). Panel C: 1BRneo cells were transfected with scrambled siRNA or
siRNA duplexes targeted to either HDAC1 or HDAC2, treated as in (A) and
stained for H2AX and HDAC1 or HDAC2. H2AX foci were enumerated as in (A).
Panel D: 1BRneo cells were transfected with scrambled siRNA or siRNA
duplexes targeted to HDAC1 and HDAC2 siRNA as in (A). Cells were treated as
in (A) and stained for γH2AX and HDAC1 or HDAC2. γH2AX foci were
enumerated as in (A). Panel E: WT or ATM-/-MEFs were transfected with KAP-1
‘C’ siRNA or scrambled siRNA (mock), using Metafectene®-Pro. 48 hr post
transfection, cells were treated and enumerated as in (A). Panel F: WT MEFs
were transfected with scrambled (control) siRNA, ATM siRNA and/or KAP-1
siRNA. 48 hr post transfection, cells were irradiated as indicated and processed
for PFGE. FAR values represent the fraction of DNA released from the gel plug
(mean and SD of 3 exp’ts). Panel G: ATM-/-MEFs were treated as in F.
Figure 6: KAP-1 association with nuclease-resistant chromatin changes
after IR. Panel A: Confluent, stationary-phase NIH 3T3 cells were exposed to 0
or 40 Gy IR and harvested 1 hr later. Cells were processed for chromatin
segregation as described in Materials and Methods. Samples were blotted for
HMGB1, KAP1, HP1, γH2AX and HDAC1 as indicated. Ponceau-S stains of
Histones are also shown. Note: While anti-KAP-1 signal decreased in C3 after
IR, antibody affinity was not affected by KAP-1 phosphorylation in whole cell
extracts (data not shown). When over-exposed, weak KAP-1 signal appeared in
the P10 and/or C1 fractions after IR (data not shown). Panel B: Confluent,
stationary-phase NIH 3T3 cells were exposed to 0-80 Gy IR as indicated and
harvested 0.5 hr later. Cells treated with 80 Gy IR were additionally harvested at
indicated times. Samples were processed as in (A) and the C3 fraction was
blotted as in (A). Panel C: DMSO or ATMi was added to confluent, stationary-
phase NIH 3T3 cells and, 0.5 hr later, cells were irradiated as indicated and
harvested 0.5 hr later. Samples were processed and the C3 fraction was blotted
as in (A).
30 Download full-text